Fixation of natural furanoeremophilane by Diels-Alder reaction

Article abstract of DOI:10.1246/bcsj.80.966

Diels-Alder reaction of furanoeremophilan-6β-ol (petasalbin) and its synthetic analogue, 4-hydroxy-4,5,6,7-tetra-hydrobenzofuran, was studied using TV-ethyl- and N-phenylmaleimides as the dienophiles, affording corresponding adducts as mixtures of stereoisomers. The adduct of petasalbin and maleimide was also obtained when the latter compound was added to a crude extracted solution of Petasites japonicus var. giganteus. By using this method, it was estimated that the contents of unstable furanoeremophilan-10β-ol in L. cymbulifera is more than 32% of the extract.

Full text of DOI:10.1246/bcsj.80.966

966 Bull. Chem. Soc. Jpn. Vol. 80, No. 5, 966–971 (2007)
Ó 2007 The Chemical Society of Japan
Fixation of Natural Furanoeremophilane by Diels–Alder Reaction
ꢀ
Kazunori Iida, Makoto Mitani, and Chiaki Kuroda
Department of Chemistry, Rikkyo University, Nishi-Ikebukuro, Toshima-ku, Tokyo 171-8501
Received November 16, 2006; E-mail: chkkuroda@grp.rikkyo.ne.jp
Diels–Alder reaction of furanoeremophilan-6ꢀ-ol (petasalbin) and its synthetic analogue, 4-hydroxy-4,5,6,7-tetra-
hydrobenzofuran, was studied using N-ethyl- and N-phenylmaleimides as the dienophiles, affording corresponding ad-
ducts as mixtures of stereoisomers. The adduct of petasalbin and maleimide was also obtained when the latter compound
was added to a crude extracted solution of Petasites japonicus var. giganteus. By using this method, it was estimated that
the contents of unstable furanoeremophilan-10ꢀ-ol in L. cymbulifera is more than 32% of the extract.
Furanoeremophilane compounds are found in nature in some
H
plants of Compositae tribe Senecioneae, such as Ligularia,1
Petasites,2 Farfugium,3 and Syneilesis.4 Although many com-
pounds have been isolated in pure form and their structures
have been determined, a problem is present in the handling
of these compounds because some of them are unstable. For
example, on the course of our investigation on the diversity
in Ligularia species in the Hengduan Mountains area,5 we
have found the presence of extremely unstable compound in
L. tongolensis, but the compound has not been isolated.5a
Some unstable furanoeremophilanes decompose even in
CDCl3 solution, and therefore, many NMR data have been
acquired in C6D6.1c,e,3,4 The instability of these compounds
are due to the presence of an electron-rich trisubstituted furan
ring, which reacts easily with electron deficient reagents, such
as proton. For the structure determination of these unstable
compounds, the problem is to obtain the targeted molecules
in pure form.
O
O
9
6
1
4
10
5
2
3
8
7
12
11
OH
OH
1
2
Chart 1.
At first, the reaction of model compound 2 was examined
with N-ethyl- and N-phenylmaleimide as the dienophiles
(Scheme 1). The reaction was carried out in EtOH/H2O 9:1
as the solvent at room temperature, which is a model condition
of extraction of natural products, since the extracted solutions
of natural products sometimes contain water and higher reac-
tion temperature must be avoided for the treatment of unstable
compounds. When 2 was treated with 3 molar amounts of N-
ethylmaleimide (3) under these conditions, the adduct was af-
forded in good yield as a mixture of three diastereomers. The
isomers of the products were partly separated by silica-gel col-
umn chromatography giving a mixture of exo-adducts 4a and
4b (57% yield, ratio 7:3), and an isomer of endo-adduct 4c
(41% yield). The fourth possible stereoisomer was not detect-
ed. The stereochemistry of the products was determined
A possible solution to this problem is to ‘‘fix’’ the furan ring
by conversion to some stable functional group. To this pur-
pose, we planned to use a Diels–Alder reaction for the trans-
formation of the furan ring into stable adduct, because furan
is a good diene,6 and especially, the furan ring of natural fura-
noeremophilanes is electron-rich and therefore the Diels–
Alder reaction is expected to proceed smoothly. A natural
intramolecular Diels–Alder adduct of 1-acyloxyfuranoeremo-
philane derivative has been reported, indicating that the reac-
tion proceeds without heating.7 Here, we report Diels–Alder
reaction of natural petasalbin and synthetic analogue with
N-phenyl- and N-ethyl-maleimide.
1
from H NMR spectra. Namely, 4a and 4b were determined
to be exo-adducts based on the J value between Ha and Hb
(J ¼ 0 Hz), which indicated that Hb is endo to 7-oxabicyclo-
[2.2.1]heptane ring system. The orientation of the hydroxy
group was axial for 4a and equatorial for 4b, respectively,
judged from the coupling constants of the proton attached to
the hydroxy-bearing carbon. However, since two conformers
were possible for the cyclohexane ring, it was difficult to deter-
mine the relative stereochemistry of 4a and 4b. Compound
4c was judged to be endo-adduct based on a relatively large
J value (5.5 Hz) between Ha and Hb.
The reaction of 2 with N-phenylmaleimide (5) was also car-
ried out similarly. Formation of adduct 6 was observed on
TLC; however, in contrast to 4a–4c, it was difficult to obtain
each products in pure form, because a retro-Diels–Alder reac-
tion occurred during the isolation process. Some other dieno-
Results and Discussion
In the present study, two substrates, furanoeremophilan-6ꢀ-
ol (=ligularol or petasalbin, 1)8,9 and a synthetic analogue 2
were used. Compound 1 was isolated from the rhizome of
the cultivated plant Petasites japonicus var. giganteus (Com-
positae),10 which was collected in Nagano Prefecture, Japan.
The synthetic analogue 2 was prepared by the Feist–Benary
method11 from 1,3-cyclohexanedione and chloroacetaldehyde
followed by LiAlH4 reduction, as described previously
(Chart 1).8
Published on the web May 14, 2007; doi:10.1246/bcsj.80.966

K. Iida et al.
Bull. Chem. Soc. Jpn. Vol. 80, No. 5 (2007)
967
Table 1. Reaction of 7 and 5 under Various Conditionsa)
O
O
2
+
N
Entry
Equiv. of 5
Time/d
Yield/%
Et
1
2
3
4
5
6
1
1
2
2
3
3
1
3
1
3
1
3
55
78
69
80
84
80
3
Et
Et
O
Hc
O
Hc
a) All reactions were carried out in EtOH/H2O (9:1) with
0.1 mol Lꢁ1 concentration of diene at room temperature.
N
N
O
Hb
Ha
O
Hb
Ha
O
O
adducts were obtained as a mixture of two stereoisomers,
and the each diastereomer was stable enough for the isolation
procedure to obtain 8a (51%) and 8b (14%) after purification
on silica gel. Both products were determined to be exo-adducts
as described above.
OH
OH
4a,4b
4c
Optimization of the reaction condition was made for the re-
action of 7 to 8. The results are shown in Table 1. The use of a
stoichiometric amount of maleimide 5 was found to be enough
if the reaction was carried out for a longer time (Entry 2).
However, in the present study, we chose to use 3 equivalent
of maleimide (Entry 5) in order to use natural product more
effectively.
O
O
2
+
N
Ph
5
Following the succession with the model compound, the
Diels–Alder reaction of natural furanoeremophilane was
studied. When compound 1 was treated with 3 in EtOH/
H2O (9:1) at room temperature, a pair of diastereomers 9a
and 9b were obtained in 39 and 16% yields, respectively
(Scheme 3). In contrast to the case of model compound 2,
the adducts with N-phenylmaleimide (5) were obtained in
better yield (74%) without the occurrence of the retro-Diels–
Alder reaction. The ratio of the two diastereomers 10a and
10b was 4:1. In both cases, stereoisomers of the two products
were obtained, and each isomer could be isolated by silica-
gel column chromatography. The stereochemistry of the all
four products were determined to be exo-adducts based on
1H NMR spectra as described for the model compounds. The
major isomer (9a/10a) and the minor isomer (9b/10b) were
determined to be the ꢀ-exo and the ꢁ-exo adducts, respective-
ly, based on the presence or absence of long-range coupling
between olefinic methyl group and the proton attached to the
hydroxy-bearing carbon.12 These protons were almost planar
with the C=C double bond for 9a/10a (J ¼ 0 Hz) but not
for 9b/10b (J ¼ 2:2 Hz).
Ph
O
N
H
O
O
H
OH
6
Scheme 1. Reaction conditions: in EtOH/H2O (9:1), r.t., 1 day.
O
i
2
OBz
ii
7
Ph
O
N
H
O
O
H
OBz
The stereoselective formation of the exo-isomers can be ra-
tionalized as follows. Although four approaching directions of
dienophile 3 (or 5) to 1 are possible, as shown in Scheme 4, ꢀ-
endo and ꢁ-endo approaches are unfavorable because of steric
congestion. Namely, the R group on the maleimide comes
close to the C-6 substituent in the ꢀ-endo approach. In the
ꢁ-endo approach, maleimide molecule must come inside the
‘‘umbrella’’ shape of the cis-decaline system of the furanoere-
mophilane. Among two exo-approaches, ꢀ-exo is more favor-
able than ꢁ-exo because steric interaction with decaline ring is
minimum. The conformation of furanoeremophilane illustrated
in Scheme 4 is ‘‘steroidal,’’ and there is another conformer,
‘‘non-steroidal.’’13 The story described here must be the same
for the non-steroidal conformation, except that steric interac-
8a,8b
Scheme 2. Reagents and conditions: i) benzoyl chloride,
pyridine, CH2Cl2, r.t., 2 h. ii) see Table 1.
philes, such as benzoquinone, ethyl acrylate, diethyl fumarate,
maleic anhydride, and diethyl acetylenedicarboxylate, were
tested other than maleimide. However, none of them afforded
the corresponding adducts in better yields than N-ethylmale-
imide.
In order to show the utility of N-phenylmaleimide, the reac-
tion was further studied using benzoate 7 as the substrate,
which was prepared by benzoylation of 2 (Scheme 2). The

968 Bull. Chem. Soc. Jpn. Vol. 80, No. 5 (2007)
Fixation of Furanoeremophilanes
1
+
3
R
O
O
N
H
OH
O
9a
(or 10a)
Me
Me
Et
H
Et
H
O
O
Me
N
N
H
H
β-exo
H
H
O
O
O
O
H
R
O
OH
O
OH
OH
9b
9b
(or 10b)
N
Me
O
9a
Me
Me
α-exo
1
+
5
O
OH
H
N
R
O
O
Me
Me
Me
Ph
Ph
O
O
β-endo
N
N
H
H
H
H
O
O
O
O
H
H
H
OH
O
Me
OH
OH
10b
O
R
N
10a
O
Me
Me
Scheme 3. Reaction conditions: in EtOH/H2O (9:1), r.t., 1 day.
α-endo
tion between C-5 methyl and dienophile must be considered
for the ꢀ-endo approach. For the reaction of the model com-
pound, related explanation can be applied for the formation
of 4a, 4b, 8a, and 8b. The formation of ꢁ-endo product (4c)
can be explained by the lack of interaction with the fused cy-
clohexane ring.
Scheme 4. Stereochemistry of the reaction of 1 and 3 (or 5).
We have recently studied chemical constituents of L.
cymbulifera, and four furanoeremophilanes have been isolated
from the plant.5a Among them, furanoeremophilan-10ꢀ-ol
(11) has been detected as the major component on TLC of
the crude extracted solution, but the compound is too labile.
From 1.5 g of the extract, only 176.3 mg (12%) of the com-
pound has been obtained. Although we have not stated in
the report, the compound decomposes when the solvent is
evaporated after purification. Ehrlich’s reaction of 11 was fast-
er than the other furanoeremophilanes, indicating that the furan
ring in 11 is highly reactive. Then, the present Diels–Alder
method was applied to ‘‘fix’’ the unstable molecule. When
N-ethylmaleimide was added to a crude extracted solution of
L. cymbulifera, the reaction immediately occurred, and the
spot of 11 on TLC disappeared. The other Ehrlich-positive
spot did not disappear in a short reaction time, indicating
that only 11 had been reacted. Form 10 cm3 of the extracted
solution, the adduct (119.1 mg) was obtained as a mixture of
two diastereomers 12a and 12b (Scheme 5). Since 2 cm3 of
the extracted solution gave 48.7 mg of the residual oil, the
result implies that the content of 11 is more than 32% of the
extract. Interestingly, minor isomer 12b was not ꢁ-exo-adduct,
but ꢀ-endo-adduct, which was determined from the J value
The present Diels–Alder reaction was applied to transform
natural furanoeremophilane into stable adduct before the isola-
tion procedure. At first, the utility of the present ‘‘fixation’’
method was estimated by comparison with the standard isola-
tion using the same extracted solution. When 12 cm3 of AcOEt
extracted solution of half-dried roots of P. japonicus var.
giganteus was treated with N-ethylmaleimide (3), 6.6 mg of
the adducts 9a and 9b (correspond to 4.3 mg of 1) were isolat-
ed. The same treatment with N-phenylmaleimide (5) afforded
8.4 mg (correspond to 4.8 mg of 1) of 10a and 10b. In contrast,
only 1.5 mg of 1 could be isolated by direct column chroma-
tography from the same amount of the extracted AcOEt solu-
tion, although acidic conditions were avoided by using neutral
silica gel. In order to reduce handling time after harvesting, the
method was also applied to ‘‘fix’’ compound 1 in fresh root.
When 3 was added to a crude extracted ethanol solution of
the fresh root, a spot of the adduct 9 immediately appeared
on TLC; however, it was difficult to judge the disappearance
of the spot of 1. From 12 g of the fresh root, 9a (54.5 mg)
and 9b (19.4 mg) were afforded as crystals.

K. Iida et al.
Bull. Chem. Soc. Jpn. Vol. 80, No. 5 (2007)
969
Wakogel C-200 was used for column chromatography unless
otherwise noted.
OH
O
+
3
Materials. Compound 1 was isolated from roots of Petasites
japonicus var. giganteus (Compositae) collected in Nagano Pre-
fecture, Japan. See our previous report8 for the isolation proce-
dure. Compound 2 was prepared via the known Feist–Benary pro-
cedure.8,11
11
in L. cymbulifera
4-Benzoyloxy-4,5,6,7-tetrahydrobenzofuran (7). In a 50 cm3
round-bottomed flask attached with CaCl2 drying tube, a stirred
solution of 2 (1.90 g, 13.8 mmol) in CH2Cl2 (50 cm3) was pre-
pared. To this was added pyridine (5.8 cm3) and benzoyl chlo-
ride (5.00 cm3, 61.8 mmol) successively, and the stirring was
continued at room temperature for 2 h. Water (40 cm3) was added,
and the mixture was extracted with CH2Cl2 and dried over an-
hydrous MgSO4. Evaporation of the solvent, followed by silica-
gel column chromatography using hexane/AcOEt (99:1) as elu-
ent, afforded 7 (3.19 g, 96%) as an oil; IR (neat) 2949, 1714,
Et
O
Et
H
O
N
N
H
H
OH
OH
O
O
O
H
O
1271, 1109, and 712 cmꢁ1 1H NMR (C6D6) ꢂ 1.30–1.40 (1H,
;
12b
12a
m), 1.49–1.58 (1H, m), 1.69–1.84 (2H, m), 2.12–2.21 (1H, m),
2.33–2.41 (1H, m), 6.06 (1H, t, J ¼ 3:8 Hz), 6.44 (1H, d, J ¼
1:5 Hz), 6.95–7.13 (4H, m), and 8.14–8.18 (2H, m); 13C NMR
(CDCl3) ꢂ 19.01, 22.76, 28.97, 66.71, 110.08, 116.42, 128.09
(2C), 129.41 (2C), 130.46, 132.62, 140.77, 153.84, and 166.09;
MS (EI) m=z 137 (Mþ ꢁ PhCO), 120 (base), 105, 91, and 77;
HRMS (EI) Found: m=z 242.0943 (Mþ); Calcd for C15H14O3:
M, 242.0943.
Scheme 5. Diels–Alder reaction of furanoeremophilan-10ꢀ-
ol in L. cymbulifera.
H
O
O
N
R
O
O
12b
Me
Diels–Alder Reaction of 2 and 3. Compound 2 (137.1 mg,
0.993 mmol) was dissolved in EtOH/H2O (10 cm3, ratio 9:1),
and to this was added a solution of N-ethylmaleimide (3, 378.9
mg, 3.03 mmol) in EtOH/H2O of the same ratio (10 cm3) at
room temperature with stirring, which was continued for 1 day.
Without extraction, water was eliminated from the mixture by
adding Na2SO4, followed by filtration. Evaporation of the sol-
vent afforded an oily residue, which was chromatographed on
silica gel (35 g) using hexane/AcOEt (8:2 to 2:8) as the eluent
to afford a mixture of 4a and 4b (148.5 mg, 57%) and 4c
(106.3 mg, 41%).
Me
Me
β-endo
Scheme 6. Stereochemistry of the formation of 12b.
(5.5 Hz) between the proton attached to the ethereal oxygen-
bearing carbon and the adjacent proton, as in the case of 4c.
This unexpected stereoselectivity can be rationalized by the
formation of hydrogen-bond between hydroxy group at C-10
and the carbonyl oxygen of the maleimide for the ꢀ-endo
approach (Scheme 6).
A mixture of 4a and 4b: Mp 138–140 ꢂC; IR (KBr) 3377 (OH),
1
1695 (C=O), 1408, 1344, 1232, and 1142 cmꢁ1; H NMR ꢂ 1.14
In conclusion, the Diels–Alder reaction of natural furano-
eremophilane compound with N-ethyl- and N-phenyl-male-
imide was found to proceed smoothly at room temperature;
however, N-ethylmaleimide was found to be better for the
model compound. Although the occurrence of retro-Diels–
Alder reaction in some case indicates limited usefulness, the
method was shown to be useful for fixation of unstable natural
furanoeremophilanes.
(3H, t, J ¼ 7:2 Hz, NCH2CH3), 1.40–2.21 (6H, m), 2.51–2.63
(1H, m), 2.72 (1H ꢃ 3/10, d, J ¼ 6:5 Hz, Hc of 4b), 2.98 (1H ꢃ
7/10, d, J ¼ 6:5 Hz, Hc of 4a), 3.00 (1H ꢃ 3/10, d, J ¼ 6:5 Hz,
Hb of 4b), 3.11 (1H ꢃ 7/10, d, J ¼ 6:5 Hz, Hb of 4a), 3.51 (2H,
q, J ¼ 7:2 Hz, NCH2CH3), 4.20 (1H ꢃ 3/10, ddd, J ¼ 2:5, 5.5,
11.6 Hz, CHOH of 4b), 4.71 (1H ꢃ 7/10, t, J ¼ 2:5 Hz, CHOH
of 4a), 5.13 (1H ꢃ 7/10, d, J ¼ 1:6 Hz, Ha of 4a), 5.14 (1H ꢃ
3/10, d, J ¼ 1:6 Hz, Ha of 4b), 6.30 (1H ꢃ 3/10, br s, C=CH
of 4b), and 6.32 (1H ꢃ 7/10, s, C=CH of 4a); 13C NMR assigned
for 4a: ꢂ 12.91, 16.50, 26.21, 32.45, 33.74, 47.63, 50.93, 64.26,
79.59, 87.80, 131.36, 149.99, 174.87, and 176.18; assigned for
4b: ꢂ 12.94, 20.41, 25.74, 33.81, 34.73, 46.76, 51.80, 67.98,
79.88, 89.58, 127.47, 153.30, 174.70, and 175.91; MS (CI)
m=z 264 (Mþ þ H), 246, 166, 138, and 121 (base); HRMS (CI)
Found: m=z 264.1235 (Mþ þ H); Calcd for C14H18NO4: M,
264.1236.
Experimental
General Procedure.
Melting points were measured on a
Laboratory Devices Mel-Temp apparatus. IR spectra were record-
ed on a Jasco FT/IR-230 spectrometer. Both 1H and 13C NMR
spectra were measured on a Jeol GSX-400 (400 MHz for 1H;
100 MHz for 13C) spectrometer in CDCl3 as the solvent, unless
otherwise noted. Chemical shifts were recorded on the ꢂ
scale (ppm) with tetramethylsilane as an internal standard. For
13C NMR, the signal of the solvent (CDCl3: 77.0 ppm) was
used as the reference. Both low-resolution mass spectra (MS)
and high-resolution mass spectra (HRMS) were obtained on a
Jeol SX-102A, CMATE II, JMS AX-500, or Shimadzu GCMS-
QP5050 mass spectrometer. Analytical TLC was done on pre-
coated TLC plates (Kieselgel 60 F254, layer thickness 0.2 mm).
4c: Mp 111–112 ꢂC; IR (KBr) 3438 (OH), 1697 (C=O), 1402,
1342, 1225, 1066, and 895 cmꢁ1
;
1H NMR ꢂ 1.05 (3H, t, J ¼
7:2 Hz, NCH2CH3), 1.34–2.43 (7H, m), 3.20 (1H, d, J ¼ 7:4 Hz,
Hc), 3.39 (2H, q, J ¼ 7:2 Hz, NCH2CH3), 3.68 (1H, dd, J ¼
5:5, 7.4 Hz, Hb), 3.99 (1H, br, W1=2 ¼ 23 Hz, CHOH), 5.17 (1H,
dd, J ¼ 1:6, 5.5 Hz, Ha), and 6.19 (1H, br s, C=CH); 13C NMR
ꢂ 12.59, 19.76, 30.66, 33.33, 34.70, 50.02, 50.98, 68.27, 78.01,

970 Bull. Chem. Soc. Jpn. Vol. 80, No. 5 (2007)
Fixation of Furanoeremophilanes
89.77, 125.32, 151.38, 174.90, and 174.97; MS (CI) m=z 263
(Mþ), 246, 166, 118, 126, and 121 (base); HRMS (CI) Found:
m=z 264.1237 (Mþ þ H); Calcd for C14H18NO4: M, 264.1236.
Diels–Alder Reaction of 7 and 5. By the same procedure
described above, compound 7 (50.5 mg, 0.209 mmol) and 5
(36.2 mg, 0.209 mmol) in EtOH/H2O afforded 8a (44.5 mg,
51%) and 8b (12.4 mg, 14%) after silica-gel chromatography us-
ing hexane/AcOEt (2:1) as the eluent.
2.01 (3H, d, J ¼ 2:2 Hz, C=CMe), 2.25 (1H, dd, J ¼ 2:2, 13.8
Hz), 2.75 (1H, d, J ¼ 6:5 Hz, CHC=O), 2.92 (1H, d, J ¼ 6:5 Hz,
CHC=O), 3.50 (2H, q, J ¼ 7:2 Hz, NCH2CH3), 4.78 (1H, br,
W1=2 ¼ 14 Hz, CHOH), and 4.78 (1H, s, CHOC); 13C NMR ꢂ
11.05, 12.97, 14.86, 18.94, 20.78, 27.05, 28.56, 29.32, 31.89,
33.78, 37.24, 41.50, 49.98, 51.22, 70.34, 83.88, 89.77, 140.21,
142.21, 174.98, and 176.33; MS (CI) m=z 360 (Mþ þ H), 342,
235, 217, 124 (base); Analysis Found: C, 70.24; H, 8.44; N,
3.61%; Calcd for C21H29NO4: C, 70.17; H, 8.13; N, 3.90%.
10a: Mp 163–164 ꢂC; IR (KBr) 3498, 2925, 1703, 1389, and
8a: An oil; IR (neat) 1712 (C=O), 1385, 1265, 1192, and
;
737 cmꢁ1
1H NMR ꢂ 1.74–2.23 (5H, m), 2.73 (1H, br d, J ¼
13 Hz), 2.97 (1H, d, J ¼ 6:5 Hz, CHC=O), 3.07 (1H, d, J ¼
6:5 Hz, CHC=O), 5.23 (1H, d, J ¼ 1:6 Hz, CH–O), 5.97 (1H,
br s, CHOBz), 6.55 (1H, br s, C=CH), 7.21–7.56 and 7.95–
7.99 (10H, m, Ph ꢃ 2); 13C NMR ꢂ 17.56, 25.98, 29.97, 47.49,
50.43, 66.74, 80.20, 88.11, 126.44 (2C), 128.56 (2C), 128.60,
129.00 (2C), 129.36 (2C), 129.85, 131.61, 133.26, 135.23,
145.91, 165.02, 173.79, and 174.88; MS (FAB) m=z 438 (Mþ þ
Na), 416 (Mþ þ H), 413, 329, 307, 154 (base), and 136; HRMS
(FAB) Found: m=z 416.1485 (Mþ þ H); Calcd for C25H22NO5:
M, 416.1499.
1190 cmꢁ1
;
1H NMR ꢂ 0.70 (3H, d, J ¼ 6:8 Hz, Me), 1.09 (3H,
s, Me), 1.24–2.34 (11H, m), 1.90 (3H, s, C=CMe), 3.08 (1H, d,
J ¼ 6:5 Hz, CHC=O), 3.31 (1H, d, J ¼ 6:5 Hz, CHC=O), 4.42
(1H, s, CHOH), 5.02 (1H, s, CHOC), and 7.26–7.48 (5H, m,
Ph); 13C NMR ꢂ 10.04, 15.80, 16.56, 19.42, 25.47, 25.73, 29.50,
30.28, 32.67, 40.09, 49.92, 50.36, 68.80, 83.87, 89.29, 126.58
(2C), 128.59, 129.06 (2C), 131.80, 141.33, 141.89, 174.56, and
175.91; MS (FAB) m=z 430 (Mþ þ Na), 413, 408 (Mþ þ H),
390, 217 (base), and 124; HRMS (FAB) Found: m=z 408.2155
(Mþ þ H); Calcd for C25H30NO4: M, 408.2176.
8b: Mp 169–170 ꢂC; IR (neat) 1712 (C=O), 1385, 1265, and
;
737 cmꢁ1 1H NMR ꢂ 1.67–1.95 (3H, m), 2.08–2.16 (1H, m),
10b: Mp 111–113 ꢂC; IR (KBr) 3450, 2929, 1705, 1387, and
1190 cmꢁ1; 1H NMR ꢂ 0.93 (3H, s, Me), 1.02 (3H, d, J ¼ 7:2 Hz,
Me), 1.18–2.33 (11H, m), 2.05 (3H, d, J ¼ 2:2 Hz, C=CMe), 2.92
(1H, d, J ¼ 6:5 Hz, CHC=O), 3.10 (1H, d, J ¼ 6:5 Hz, CHC=O),
4.82 (1H, br dq, J ¼ 8:7, 1.9 Hz, CHOH), 4.92 (1H, s, CHOC),
and 7.24–7.48 (5H, m, Ph); 13C NMR ꢂ 11.15, 14.87, 18.98,
20.76, 27.11, 28.56, 29.28, 31.89, 32.23, 41.51, 50.04, 51.34,
70.37, 84.36, 90.27, 126.53 (2C), 128.65, 129.09 (2C), 131.72,
140.43, 142.41, 174.17, and 175.64; MS (FAB) m=z 430 (Mþ þ
Na), 413, 408 (Mþ þ H), 390, 329, 307, 154 (base), and 136;
HRMS (FAB) Found: m=z 408.2151 (Mþ þ H); Calcd for
C25H30NO4: M, 408.2176.
2.29–2.36 (1H, m), 2.69 (1H, br d, J ¼ 13 Hz), 3.01 (1H, d, J ¼
6:5 Hz, CHC=O), 3.19 (1H, d, J ¼ 6:5 Hz, CHC=O), 5.28 (1H,
d, J ¼ 1:5 Hz, CH–O), 5.65 (1H, ddd, J ¼ 2:6, 5.9, 11.2 Hz,
CHOBz), 6.26 (1H, br s, C=CH), 7.25–7.64 and 8.05–8.09
(10H, m, Ph ꢃ 2); 13C NMR ꢂ 20.33, 25.76, 30.76, 46.93,
51.61, 69.25, 80.48, 90.19, 126.51 (2C), 128.51 (2C), 128.58,
128.72, 129.09 (2C), 129.57, 129.71 (2C), 131.65, 133.41, 148.96,
165.72, 173.72, and 175.04; MS (FAB) m=z 438 (Mþ þ Na), 416
(Mþ þ H), 413, 391, 294, 154 (base), and 121; HRMS (FAB)
Found: m=z 416.1482 (Mþ þ H); Calcd for C25H22NO5: M,
416.1499.
Diels–Alder Reaction of Unpurified Extract of Petasites
Diels–Alder Reaction of Furanoeremophilan-6ꢀ-ol (1). In a
10 cm3 round-bottomed flask, a solution of compound 1 (13.1
mg, 0.056 mmol) in EtOH/H2O (0.3 cm3, ratio 9:1) was prepared,
and to this was added a solution of 5 (30.3 mg, 0.175 mmol) in the
same solvent (0.3 cm3) with stirring at room temperature. After
being stirred for 24 h, water was removed by adding anhydrous
Na2SO4 followed by filtration. The solvent EtOH was evaporated,
and the resultant residue was chromatographed on silica gel (3 g)
using hexane/AcOEt (4:1 to 0:1) as the eluent to afford a mixture
of 10a and 10b (16.8 mg, 74%). Similarly, the reaction of 1
(12.5 mg, 0.0533 mmol) and 3 (23.0 mg, 0.184 mmol) in EtOH/
H2O (0.5 cm3) afforded 9a (7.7 mg, 39%) and 9b (3.2 mg, 16%)
after silica-gel (2 g) column chromatography using hexane/AcOEt
(7:3 to 4:6) as eluent.
japonicus var. giganteus.
Method and the Conventional Isolation:
Comparison of the Diels–Alder
Half-dried roots
(569 g) were extracted with AcOEt at room temperature as usual.
The extracted solution (36 cm3) was obtained by decantation
and was divided into three portions. To the first 12 cm3 portion
was added a solution of N-ethylmaleimide (126.8 mg) in AcOEt
(5 cm5), and the mixture was stirred at room temperature for 1
week. Evaporation of the solvent followed by silica-gel (2 g) col-
umn chromatography using hexane/AcOEt (9:1) as the eluent af-
forded a mixture of 9a and 9b (6.6 mg). The second 12 cm3 por-
tion of the extracted solution was similarly treated with a solution
of N-phenylmaleimide (175.7 mg) in AcOEt (5 cm3) giving 10a
and 10b (8.4 mg). The third 12 cm3 portion was evaporated to
an oily residue (52.4 mg), which was chromatographed on neutral
silica gel (Silica Gel 60 N) (5 g) using hexane/AcOEt (9:1) as the
eluent to afford 1 (1.5 mg).
9a: Mp 70–72 ꢂC; IR (KBr) 3448, 2927, 1693, 1406, and
1
999 cmꢁ1; H NMR ꢂ 0.69 (3H, d, J ¼ 6:5 Hz, Me), 1.08 (3H, s,
Me), 1.14 (3H, t, J ¼ 7:2 Hz, NCH2CH3), 1.23–1.94 (8H, m), 1.86
(3H, s, C=CMe), 2.22–2.29 (3H, m), 2.91 (1H, d, J ¼ 6:5 Hz,
CHC=O), 3.13 (1H, d, J ¼ 6:5 Hz, CHC=O), 3.51 (2H, q, J ¼
7:2 Hz, NCH2CH3), 4.39 (1H, br s, CHOH), and 4.89 (1H, s,
CHOC); 13C NMR ꢂ 9.92, 12.97, 15.76, 16.53, 19.44, 25.50,
25.68, 29.50, 30.29, 32.69, 33.71, 40.09, 49.83, 50.29, 68.77,
83.39, 88.79, 140.20, 141.67, 175.37, and 176.60; MS (CI) m=z
360 (Mþ þ H), 342, 234, 217 (base), 126, and 124; HRMS (CI)
Found: m=z 360.2172 (Mþ þ H); Calcd for C21H30NO4: M,
360.2176.
Application to the Extraction of Fresh Root: Fresh roots
(12 g) were extracted with ethanol (13 cm3) for 1 week at room
temperature. The plant material was filtered off, and to the filtrate
was added N-ethylmaleimide (108.7 mg). The reaction mixture
was stirred at room temperature for 1 week, followed by evapora-
tion of the solvent. The resultant residue was chromatographed as
described above to afford 9a (54.5 mg) and 9b (19.4 mg).
Diels–Alder Reaction of Unpurified Extract of Ligularia
cymbulifera.
Dried roots, collected near Zhongdian city,
Yunnan, China, were extracted with ethyl acetate for 1 week at
room temperature. An extracted solution (10 cm3) was obtained by
decantation, and a solution of N-ethylmaleimide (2.8 g) in AcOEt
(4 cm3) was added. After keeping the mixture for 30 min at room
9b: Mp 151–152 ꢂC; IR (KBr) 3448, 2925, 1697, and 1400
cmꢁ1
;
1H NMR ꢂ 0.91 (3H, s, Me), 1.11 (3H, d, J ¼ 7:1 Hz,
Me), 1.13 (3H, t, J ¼ 7:2 Hz, NCH2CH3), 1.19–2.10 (10H, m),

K. Iida et al.
Bull. Chem. Soc. Jpn. Vol. 80, No. 5 (2007)
971
temperature, the solvent was evaporated off, and the residual oil
was separated by silica-gel (8 g) column chromatography using
hexane–AcOEt (9:1 to 3:1) as the eluent to afford a mixture of
12a and 12b (119.1 mg, ratio 3:2). The following spectral data
were obtained after further separation.
At the same time, in order to calculate the content of 11, the
extracted solution of L. cymbulifera (2 cm3) was simply evaporat-
ed to yield an oily residue (48.7 mg).
References
For examples: a) F. Patil, G. Ourisson, Y. Tanahashi,
1
M. Wada, T. Takahashi, Bull. Soc. Chim. Fr. 1968, 1047. b) Y.
Ishizaki, Y. Tanahashi, T. Takahashi, K. Tori, J. Chem. Soc. D
1969, 551. c) M. Tada, Y. Moriyama, Y. Tanahashi, T. Takahashi,
Bull. Chem. Soc. Jpn. 1974, 47, 1999. d) T. Sato, Y. Moriyama,
H. Nagano, Y. Tanahashi, T. Takahashi, Bull. Chem. Soc. Jpn.
1975, 48, 112. e) F. Bohlmann, C. Zdero, Phytochemistry 1980,
19, 1550. f) F. Bohlmann, M. Grenz, Phytochemistry 1979, 18,
491.
2 For examples: a) K. Naya, M. Nakagawa, M. Hayashi,
K. Tsuji, M. Naito, Tetrahedron Lett. 1971, 12, 2961. b) L.
´
Novotny, K. Kotva, J. Toman, V. Herout, Phytochemistry 1972,
11, 2795.
12a: Mp 138–139 ꢂC; IR (KBr) 3517 and 1682 cmꢁ1; 1H NMR
ꢂ 0.77 (3H, d, J ¼ 6:5 Hz, Me), 0.97 (3H, s, Me), 1.15 (3H, t,
J ¼ 7:1 Hz, NCH2CH3), 1.25–1.97 (8H, m), 1.74 (3H, d, J ¼
2:5 Hz, C=CMe), 2.09 (1H, d, J ¼ 14:6 Hz), 2.10 (1H, dq, J ¼
16:6, 2.5 Hz), 2.33 (1H, d, J ¼ 16:6 Hz), 2.63 (1H, d, J ¼ 14:6
Hz), 2.86 (1H, d, J ¼ 6:4 Hz, CHC=O), 2.90 (1H, d, J ¼ 6:4 Hz,
CHC=O), 3.51 (2H, q, J ¼ 7:1 Hz, NCH2CH3), and 4.88 (1H, s,
CHOC); 13C NMR ꢂ 9.87, 12.98, 14.35, 15.71, 22.61, 28.62,
29.57, 33.33, 33.45, 33.72, 36.50, 41.36, 49.82, 50.93, 74.52,
80.07, 88.63, 135.89, 139.60, 176.30, and 176.62; MS (CI) m=z
360 (Mþ þ H), 342, 291, 235, 217 (base), 126, and 89; HRMS
(CI) Found: m=z 360.2172 (Mþ þ H); Calcd for C21H30NO4: M,
360.2176.
12b: Mp 129–131 ꢂC; IR (KBr) 3479 and 1682 cmꢁ1; 1H NMR
ꢂ 0.72 (3H, d, J ¼ 6:5 Hz, Me), 0.97 (3H, s, Me), 1.02 (3H, t, J ¼
7:1 Hz, NCH2CH3), 1.11–1.90 (8H, m), 1.69 (3H, s, C=CMe),
2.14 (1H, d, J ¼ 15:4 Hz), 2.30 (1H, br d, J ¼ 16:5 Hz), 2.87
(1H, d, J ¼ 15:4 Hz), 3.27 (1H, d, J ¼ 7:3 Hz, CHC=O), 3.41
(2H, q, J ¼ 7:1 Hz, NCH2CH3), 3.69 (1H, dd, J ¼ 5:5, 7.3 Hz,
CHC=O), 4.53 (1H, br, OH), and 4.95 (1H, d, J ¼ 5:5 Hz,
CHOC); 13C NMR ꢂ 11.09, 13.13, 14.58, 15.60, 22.49, 29.34,
29.70, 33.48, 33.53, 33.88, 38.56, 42.89, 49.55, 52.49, 73.87,
81.62, 88.79, 134.93, 137.23, 174.31, and 177.37; MS (CI) m=z
360 (Mþ þ H), 342, 291, 235, 217 (base), and 126; HRMS (CI)
Found: m=z 360.2184 (Mþ þ H); Calcd for C21H30NO4: M,
360.2176.
3
H. Nagano, Y. Tanahashi, Y. Moriyama, T. Takahashi,
Bull. Chem. Soc. Jpn. 1973, 46, 2840.
C. Kuroda, T. Murae, M. Tada, H. Nagano, T. Tsuyuki, T.
Takahashi, Bull. Chem. Soc. Jpn. 1982, 55, 1228.
a) R. Hanai, X. Gong, M. Tori, S. Kondo, K. Otose, Y.
4
5
Okamoto, T. Nishihama, A. Murota, Y. Shen, S. Wu, C. Kuroda,
Bull. Chem. Soc. Jpn. 2005, 78, 1302. b) H. Nagano, Y. Iwazaki,
X. Gong, Y. Shen, C. Kuroda, R. Hanai, Bull. Chem. Soc. Jpn.
2006, 79, 300. c) M. Tori, K. Honda, H. Nakamizo, Y. Okamoto,
M. Sakaoku, S. Takaoka, X. Gong, Y. Shen, C. Kuroda, R. Hanai,
Tetrahedron 2006, 62, 4988.
6
W. Oppolzer, Comprehensive Organic Synthesis, ed. by
B. M. Trost, Pergamon, Oxford, 1991, Vol. 5, pp. 315–399.
Y. Zhao, S. Parsons, R. J. Baxter, Z.-J. Jia, H.-D. Sun,
D. W. H. Rankin, Chem. Commun. 1996, 2473.
C. Kuroda, T. Ueshino, H. Nagano, Bull. Chem. Soc. Jpn.
2004, 77, 1737.
7
8
9 a) H. Ishii, T. Tozyo, H. Minato, Tetrahedron 1965, 21,
2605. b) K. Yamakawa, T. Satoh, Chem. Pharm. Bull. 1979, 27,
1747.
´
10 L. Novotny, K. Kotva, J. Toman, V. Herout, Phytochem-
istry 1972, 11, 2795.
The authors must thank Prof. M. Tori and Ms. Y. Okamoto
of Tokushima-bunri University for the measurements of
the mass spectra. This work was conducted as part of Frontier
Project ‘‘Adaptation and Evolution of Extremophile’’ from the
Ministry of Education, Culture, Sports, Science and Technolo-
gy of Japan. Financial support was also obtained from JSPS
(Grant-in-Aid for Scientific Research No. 16404008).
11 E. Bisagni, J.-P. Marquet, J.-D. Bourzat, J.-J. Pepin,
´
J. Andre-Louisfert, Bull. Soc. Chim. Fr. 1971, 4041.
12 Atta-ur-Rahman, Nuclear Magnetic Resonance, Springer-
Verlag, New York, 1986.
13 T. Sato, M. Tada, T. Takahashi, Tetrahedron Lett. 1977,
18, 3895.